Wide-field-of-view static lidar

Abstract

The invention relates to a method (100) for processing a signal from a lidar, the method comprising the steps of: A: digitising the amplified electrical signal (formula I); B: applying at least one time correction function, referred to as the correcting filter (Ce(t)), to the digitising amplified electrical signal in order to generate a processed signal (sf(t)), the correcting filter (Ce(t)) being determined based on the impulse response and a predetermined time analysis function, the analysis function having at least one non-zero value (a0, a1, a2), referred to as the discontinuity, at a given time referred to as the discontinuity time (td0, td1, td2), with a return to substantially zero values around the discontinuity; C: determining a distance (di) of the at least one element (Ei) based on the processed signal.

Description

DOMAINE DE L'INVENTION

[0001] The present invention relates to the field of lidars performing time-of-flight (TOF) measurements, and more particularly to lidars with a field of view greater than 5°. This focuses on static lidars (without moving mechanical parts) that are low-power, small in size, and low-cost. This type of lidar finds applications, for example, in obstacle detection. ETAT DE LA TECHNIQUE

[0002] A lidar (Light Detection And Ranging) is a device used to measure distance by measuring the time of flight of a light pulse.

[0003] To do this, it emits a high-power, short-duration light pulse (typically a few nanoseconds), and later receives a pulse reflected / backscattered from an obstacle. Knowing the speed of light, the distance is deduced from this delay, called the time of flight. It is calculated using the formula d = c . R 2 in which c = 3.10 8< m / s is the speed of light, R the delay due to distance d between the transmitter and the obstacle. Thus, for an obstacle 1m away, it is assumed that the delay is 6.67 ns.

[0004] Lidar consists of: from an optical pulse emitter (laser diode or light-emitting diode), controlled by a logic component such as a microprocessor, microcontroller, or FPGA; from an optoelectronic receiver whose role is to convert the reflected pulse into an electrical signal while maintaining measurement quality, in particular by maximizing the measurement signal-to-noise ratio (S / N); from a processing unit of the electrical signal supplied by the receiver to deduce the distance to the obstacle. This unit can be a time-distance converter, or « Time to Distance Convert » (TDC), or a digital signal processing system based on a microprocessor, microcontroller or FPGA.

[0005] Typically, when using a laser diode emitter, the emitted optical power can reach a few tens of watts over a period of a few nanoseconds. The orders of magnitude of the received power from the echoes are typically from a few nanowatts to a few hundred milliwatts. The background light can vary from zero power in complete darkness up to 1 kW / m² (120 klux) in full sunlight. Thus, an irradiance of 10 W / m², corresponding to average artificial lighting illuminating a 5 x 5 mm² silicon photodiode, will generate a photogeneration current of approximately 500 µA , full sun a current of about 20 mA .

[0006] Lidars are used to measure distances, map surfaces, and detect objects. There are several types of lidars:

[0007] 1D lidar, often with a narrow field of view, features a pulsed light beam emitted from a small aperture (low field of view lidar - Field Of View - FOV It measures the distance to a precise point where the object is located. Optoelectronic components (beam-emitting diode, receiving photodiode) are often combined with optical components (lenses, filters, etc.) for collimation and beam focusing. These narrow-field (typically less than 3°) and short-range 1D lidars are marketed for obstacle detection applications in lightweight applications involving autonomous or semi-autonomous moving objects such as drones or robots in the broadest sense (autonomous vacuum cleaners and lawnmowers, remote-controlled vehicles with obstacle detection, etc.). They have superseded traditional ultrasonic rangefinders, which have a much wider field of view and where it is impossible to know exactly which object will be the first obstacle detected.

[0008] 1D lidars have the advantage of being relatively compact and low cost but have a major disadvantage which is a narrow field of view, less than 3° (Safran LRF3013, Benewake TF02).

[0009] The need for obstacle detection requires a wide field of view, and this role is currently primarily fulfilled by the use of 3D lidars with scanning technology. The pulsed beam also has a narrow aperture, but is coupled with a mechanical scanning system to irradiate a large portion of the space: several shots directed at different locations are necessary to create a map. 3D lidars allow for precise mapping of the environment through the use of more or less complex mechanical systems or even through the use of MEMS (Micro Electro Mechanics Systems). They are effective but have the disadvantage of being oversized for light-duty applications, in terms of: Data volume: accurate mapping requires heavy data processing, and it's possible that not all of this data will need to be processed. Significant physical footprint. High power consumption (greater than 1W in most cases). Low measurement refresh rate (a few tens of Hz).

[0010] This is why, in the field of reversing radars requiring obstacle detection over a wide field of view, camera-based devices, rather than lidars, have supplanted or complemented ultrasonic sensors. Indeed, for collision avoidance, the wide field of view of an ultrasonic sensor is an advantage over traditional 3D lidar, which, to scan a wide area, needs to be mounted on a rotating platform. Furthermore, ultrasonic sensors are inherently easier to use due to the slow propagation speed of acoustic waves compared to that of electromagnetic waves, resulting in much longer echo times that are simpler to measure. These sensors are found in numerous applications, including vehicle collision avoidance, robotics, and the detection of objects or living beings.

[0011] However, the wavelength and propagation speed characteristics of acoustic waves mean that ultrasonic sensors have fundamental flaws: High sensitivity to atmospheric and environmental conditions: humidity, rain, temperature, noise; it is difficult to obtain ranges beyond one meter with reasonable sensitivity under these climatic conditions; the presence of parasitic secondary lobes in the measurement field with a risk of false detections; insensitivity to low-roughness surfaces (less than 1 mm) viewed at a high angle of incidence (total internal reflection prevents their detection); difficulty in detecting thin objects, small dimensions, or surfaces; the slow measurement speed is a limitation to the detection of obstacles moving relative to the sensor

[0012] A lidar with the wide-field properties of ultrasonic sensors, illuminating a scene with a cone larger than 5°, or even 10° or 20°, would allow for scanning a large portion of space and detecting echoes from various elements or obstacles within the scene or field of observation, with the possibility of a single pulse (eliminating the need for scanning). Furthermore, a lidar with a wide field of view, virtually unaffected by high humidity and / or rain, and capable of detecting smooth painted surfaces at high angles of incidence, would have a competitive advantage in collision avoidance applications currently reserved for ultrasonic sensors. It could also be used for the autonomous movement of robots or drones.

[0013] However, a wide field of view is not the natural domain of lidar. Indeed, the backscattered optical flux decreases with distance (d) as d - 4 for small objects, instead of decreasing as d - 2 for a narrow field, which limits its range. The difference lies in the incident light intensity density. In a so-called "narrow" field, it is assumed that all the incident energy is contained within the surface of the obstacle (in other words, the surface area of ​​the obstacle is greater than the illuminated surface along the solid angle): the obstacle receives all the energy from the emitter. Each point on the obstacle then backscatters energy towards the receiver according to a law of d - 2.In the case of a wide field, the obstacle is totally included in the illumination cone: the obstacle receives only part of the energy from the emitter according to a law in d -2< , and backscatters this energy towards the receiver also according to a law in d -2< , with an overall energy on the receiver in d -4< of the energy of the emitter.

[0014] Furthermore, large objects in the background obscure small ones in the foreground, for reasons explained later. Centimeter-level resolution is more difficult to achieve, and the costs are likely higher than for an ultrasonic sensor.

[0015] Typically, the receiver exhibits a known impulse response hr(t), which is determined by measurement or calculation. A state-of-the-art method for processing a lidar signal is to operate in Fourier space. The signal at the receiver's output is digitized, and its Fourier transform is calculated. Then, a filter F0(f) is applied, whose transfer function is close to the inverse of the Fourier transform of hr(t): F0(f) = 1 / Hr(f), where Hr(f) is the Fourier transform of hr(t).

[0016] These methods, commonly referred to as spectrum whitening through a whitening filter, are widely described in the literature but raise numerous stability and complexity issues. They most often result in the need for high computing power, beyond the capabilities of a simple, low-power embedded microprocessor. Furthermore, they pose processing problems, as the presence of noise introduces division by zero in the calculations. Patent publications US2013 / 088726A1 and WO2013 / 120041A1, as well as the publication by Parrish: "Empirical comparison of full-waveform lidar algorithms: Range extraction and discrimination performance", PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING, vol. 77, no. 8, August 1, 2011 (2011-08-01), pages 825-838, XP055049568, disclose prior art lidars.

[0017] One aim of the present invention is to overcome some of the aforementioned drawbacks by proposing a lidar signal processing method for a wide field-of-view TOF lidar, enabling discrimination of elements / obstacles present in the illuminated scene and determination of their distance, all in a single shot. DESCRIPTION DE L'INVENTION

[0018] The present invention relates to a method for processing a signal from a lidar, said lidar performing a time-of-flight measurement and comprising a transmitter configured to emit light pulses towards a scene at an angle greater than or equal to 5° and a receiver, said receiver having an impulse response and comprising a photodetector configured to receive pulses reflected or backscattered by at least one element of the scene and to convert said pulses into an electrical signal, and an amplification circuit configured to generate an amplified electrical signal, the method comprising the steps of: ∘ A: digitize said amplified electrical signal ∘ B: apply at least one time correction function, called a correction filter, to said digitized amplified electrical signal, so as to generate a processed signal, said correction filter being determined from said impulse response and a predetermined time analysis function, said analysis function having at least one non-zero value, called a discontinuity, at a given instant called the instant of discontinuity, with a return to substantially zero values ​​around said discontinuity, ∘ C: determine a distance (di) from said at least one element (Ei) to said processed signal.

[0019] According to the invention, the application of the correction filter consists of a convolution of the digitized amplified electrical signal with said time correction function, said correction filter being determined by deconvolution of said impulse response by said predetermined analysis function.

[0020] According to one embodiment, the presence of said at least one element in said scene corresponds to a local maximum of said processed signal, and said associated distance is determined from a temporal location of said local maximum.

[0021] According to one embodiment, said impulse response has a maximum at a time t m-imp, and said at least one instant of discontinuity of the analysis function coincides temporally with said instant t m-imp.

[0022] According to one embodiment, the analysis function has zero values ​​outside of said at least one discontinuity.

[0023] According to one embodiment, an analysis function presents either a single discontinuity, two discontinuities, or three discontinuities, located respectively at discontinuity instants close in time, that is to say separated by a few sampling points.

[0024] According to one embodiment, a plurality of corrective filters determined from a plurality of analysis functions are applied so as to generate a plurality of associated processed signals, said distance to said at least one element of the scene being determined from said plurality of processed signals.

[0025] According to one embodiment, the plurality of corrective filters is applied via an iterative process until a final processed signal allows the determination of a distance corresponding to the nearest obstacle; the iterative process consisting of modifying discontinuities, i.e. non-zero values ​​of the said analysis functions.

[0026] According to one embodiment, a corrective filter corresponding to an analysis function with a single discontinuity is first applied, followed by analysis functions with two or three discontinuities, the said discontinuities being modified iteratively.

[0027] According to another aspect, the invention relates to a lidar system for measuring time of flight comprising: an emitting device configured to emit light pulses towards a scene at an angle greater than or equal to 5°; a receiving device exhibiting an impulse response and comprising: ∘ a photodetector configured to receive pulses reflected or backscattered by at least one element of the scene and to convert said pulses into an electrical signal, ∘ an amplification circuit configured to amplify said electrical signal, a processing unit for said amplified electrical signal configured to: ∘ digitize said amplified electrical signal, ∘ apply at least one time-domain correction function, called a correction filter, to said digitized amplified electrical signal, so as to generate a processed signal (sf(t));the application of the correction filter consisting of a convolution of the digitized amplified electrical signal with said time correction function, said correction filter being determined from said impulse response and a predetermined time analysis function, said analysis function having at least one non-zero value, called discontinuity, at a given instant called instant of discontinuity, with a return to substantially zero values ​​around said discontinuity; said correction filter is determined by deconvolution of said impulse response by said predetermined analysis function, ∘ determine a distance of said at least one element from said processed signal. ;

[0028] According to one embodiment, the amplification circuit comprises a transimpedance amplifier, a transformer comprising a primary and a secondary, and a capacitor, the primary of the transformer being connected to an anode of the photodetector, the secondary being connected to a capacitor, said capacitor being connected to an input of said transimpedance amplifier.

[0029] The following description presents several embodiments of the device of the invention; these examples are not limiting to the scope of the invention. These embodiments illustrate both the essential features of the invention and additional features related to the embodiments considered.

[0030] The invention will be better understood, and other features, purposes, and advantages thereof will become apparent from the detailed description that follows and with reference to the accompanying drawings, which are given by way of non-limiting examples and on which: There figure 1A illustrates a lidar according to the invention. The figure 1B This illustrates a drone landing, equipped with a 1D lidar system according to the prior art. figure 1C illustrates a drone landing, equipped with a lidar according to the invention. figure 2 illustrates the evolution of the convolution for four time values, for the situation represented figure 1 . There figure 3 illustrates the signal s0(t) obtained at the receiver output. figure 4 illustrates the concept of a hypothetical obstacle. The figure 5 illustrates the temporal evolution of different signals of interest. The figure 6 illustrates the lidar signal processing method according to the invention. figure 7 illustrates the measurement and processing chain. The figure 8 illustrates the corrected impulse response. The figure 9 This illustrates the temporal method, which transforms the unprocessed impulse response of the receiver into a predetermined analysis function. figure 10 illustrates the coefficients of the correction function, calculated by deconvolving the impulse response with the predetermined analysis function. figure 11 illustrates the effect of an analysis function with a single discontinuity on a signal Ir. figure 12 This illustrates the limitation of an analysis function with a single discontinuity and a certain width on a signal Ir. figure 13 illustrates an analysis function exhibiting a single discontinuity. figure 14 illustrates an analysis function exhibiting two discontinuities. figure 15 illustrates a variant of the method according to the invention in which the corrective filter is applied iteratively. figure 16 illustrates the raw and processed signals for an initial configuration of 3 obstacles. figure 17 illustrates an example of a simple Dirac analysis function (A) and an example of a double Dirac analysis function (B). figure 18 illustrates the processed signal for a second configuration of 3 obstacles. figure 19 illustrates the processed signal for a third configuration of 3 obstacles. The figure 20 illustrates the signals processed for the third configuration, which is iteratively processed by a dual Dirac analysis function. figure 21 illustrates a lidar receiving device according to the prior art. figure 22 illustrates a receiving device according to a variant of the lidar according to the invention. DESCRIPTION DETAILLEE DE L'INVENTION

[0031] A wide FOV 10 TOF lidar according to the invention is illustrated. figure 1A It comprises a DE emission device configured to emit light pulses towards a scene at a field of view (FOV) greater than or equal to 5°, preferably 10°. The emitting element is, for example, a laser diode or a light-emitting diode. The choice of wavelengths for a lidar according to the invention is wider than for a narrow-field lidar, because using a wide field greatly reduces the risks to eye safety. Preferably, the illumination wavelength is chosen close to the maximum sensitivity of the detector in order to optimize reception.

[0032] The lidar 10 also includes a receiver device DR (or receiver) comprising a photodetector PD configured to receive pulses reflected or backscattered by at least one element (Ei i, element index) in the scene and to convert the reflected pulses into an electrical signal, and an amplification circuit CA configured to amplify the electrical signal. Typically, the receiver (photodetector and amplification circuit) exhibits an impulse response hr(t) that can be measured and / or calculated.

[0033] A processing unit (PU) controls the transmission, typically via a logic component such as a microprocessor, microcontroller, or FPGA, digitizes the amplified electrical signal, and processes it to extract useful information, namely the presence of elements in the detection field and their respective distances. The lidar processing unit (PU) according to the invention implements a particular processing method as described later.

[0034] In the example of the figure 1A The detection field contains two elements, E1 and E2, representing a post P and a vehicle V, respectively. The initial pulse li is emitted at time t=0, and the photodetector receives, at time t1, a first pulse Ir1 from the backscatter of post P and, at time t2, a second pulse Ir2 from the backscatter of vehicle V. The detector also receives ambient light, such as sunlight, if applicable. For the example of a reversing lidar according to the invention mounted on a vehicle V, Figure 1 illustrates the scene illuminated by the lidar on vehicle V. Generally, the lidar according to the invention can be mounted on any moving object: automobile, drone, robot, cane for the visually impaired, etc. For proper operation, the lidar must detect the presence of post P and determine its distance d1 without being hindered by the backscatter of vehicle V.

[0035] According to another example, Lidar is static and detects the presence of static or moving objects.

[0036] The use of a wide-field-of-view proximity lidar according to the invention as illustrated in the figure 1 This involves specific measurement characteristics and difficulties. The wide detection field means that it can contain multiple obstacles. This differs from many other lidar applications where the beamwidth of the emitters is reduced by adding specific collimation optics, resulting in only one obstacle being visible in the field of analysis.

[0037] There figure 1B illustrates a drone D in action descending on rough terrain equipped with a lidar 1D 2 according to the state of the art: obstacle detection is uncertain due to the narrow field.

[0038] There figure 1C illustrates this same drone equipped with a lidar 10 according to the invention allowing, thanks to the large field and the signal processing according to the invention, an exact detection of obstacles.

[0039] A primary consequence of the wide field of view is the spatial spread of the emitted energy, leading to reduced illumination of obstacles. These obstacles, in turn, produce weaker echoes, which are more difficult to measure than with a focused or collimated beam. This point, while also contributing to eye safety, raises the issue of the measurement signal-to-noise ratio.

[0040] A second consequence of opening up the detection field is the greater probability of finding a strong emitting source, such as the sun.

[0041] Finally, the proximity of obstacles means that echo times are measured in short intervals, which a priori requires fast detection electronics.

[0042] In the remainder of this document, we assume that the emitter provides a perfectly localized temporal light pulse of the Dirac delta function type: I i t = δ t .

[0043] An obstacle located at a distance d 1 will reflect some of the energy of this impulse, called an echo, back towards the receiver. The echo can be described by the following relationship: I r 1 t = a 1 δ t − R avec R 1 = d 1 c , where c is the speed of light, a 1. Amplitude, R 1. The delay related to distance d 1 between the transmitter and the obstacle.

[0044] If N obstacles are present in the detection field, the receiver measures an optical signal composed of the sum of the echoes: I r t = ∑ n = 1 N a n δ t − R n

[0045] In addition to these information-carrying optical signals, there is a background light. A 0 (ambient lighting, sunlight) considered here as constant or slightly variable, to arrive at the overall optical signal I gl t = A 0 + ∑ n = 1 N a n δ t − R n

[0046] In a conventional way, optical signal recovery I gl ( t ) by the receiver, then its conversion into an amplified electrical signal s 0 ( t ) by means of an amplifier system, is accompanied by certain modifications to the signal. Amplification is inevitably associated with the addition of measurement noise α(t) and a limited bandwidth. This limitation results in a temporal broadening of each reflected pulse due to the convolution effect between the pulses and the impulse response. h r ( t ) of the detector-amplifier assembly (receiver): s 0 t = I gl t + n ˜ t ∗ h r t

[0047] The direct consequence of noise is an error in the temporal localization of the return time of an echo pulse, or even the impossibility of detection if the latter is drowned out by noise.

[0048] Limiting the amplifier's bandwidth makes it difficult, or even impossible, to discriminate between echo pulses that are too close together. Indeed, the impulse response h r ( t The receiver's frequency response (detector and amplifier circuit) is directly related to its frequency response, and therefore its bandwidth. According to the results of Fourier analysis theory for linear systems (here, the receiver), a low bandwidth will lead to a long impulse response.

[0049] For better understanding, it is recalled that the convolution operator, denoted *, has the following generic mathematical expression: s 0 t = h r ∗ I r t = ∫ h r u I r t − u du .

[0050] In the following, noise and the DC component of the illuminance are not taken into account. In the formulation of equation (5), the input quantity is a luminous intensity Ir, the output quantity is a voltage s0(t), and hr(t) is the impulse response of the photodetector + amplifier assembly.

[0051] Taking the example of a scene with two echoes: I r t = I r 1 t + I r 2 t = I 1 t − R 1 + I 2 t − R 2 with I1 and I2 respectively proportional to a1 and a2 (formula (2)).

[0052] For example, the figure 2 illustrates the evolution of the convolution for four values ​​of t, t '1 to t '4 for the situation represented figure 1 , with a thin pole P reflecting little echo (Ir1) placed in front of a car V reflecting more echo (Ir2): has) t = t 1: Only the impulse Ir1 interacts with the impulse response h r .The common area between the impulse and the impulse response is small, the output signal is also small and begins a slow growth. b) t = t 2: The two impulses Ir1 and Ir2 interact with the impulse response h r . The common areas increase, and the output signal grows, through the interaction of the two impulses with the impulse response. c) t = t 3: the two impulses interact with h r in a maximal manner. The common areas are maximized, as is the output signal. d) t = t 4: The two impulses slowly emerge from the impulse response h r , First the pulse Ir1 then the pulse Ir2. The common areas decrease slowly, the output signal decreases slowly.

[0053] There figure 3 illustrates the signal s 0 (t) obtained at the receiver output.

[0054] The standard processing of the signal from the receiver to detect the presence of an obstacle is carried out by searching for the maximum and the associated time tm, assumed to correspond to the echo of an obstacle to be detected and therefore to the delay R. The distance of the obstacle is then determined with formula (1), with tm delay linked to the distance (time of flight).

[0055] THE figures 2 And 3These findings show that the output signal from the receiver is the result of interaction with only one of the two pulses at the beginning or end of the cycle, where the impulse response is most gradual, and therefore where the output signal rises or falls most slowly. Ultimately, the observed output signal s0(t) is almost always the result of the receiver's interaction with both pulses, without being able to clearly distinguish the effects of each. For example, the effect of spacing out and especially bringing together the two pulses would produce only a small difference in the output signal. Temporal discrimination is therefore not possible.

[0056] The maximum of s0(t) is a kind of barycenter of the two echo pulses as a function of their respective amplitudes and delays (ai, ti), ultimately corresponding to a fictitious obstacle, instead of two real ones: the standard processing identifies a fictitious obstacle Of at t'3, corresponding to a distance df, and not two obstacles at distances d1 and d2, as illustrated. figure 4 Without accounting for the presence of multiple echoes, this fictitious obstacle is always located further away than the nearest obstacle. For example: an obstacle is located at d 1 = 1m, and another further away located at d 2 = 3m. The imaginary obstacle will be located between d 1 ≤ obstacle fictif ≤ d 2. It is located all the further away from d 1 that the amplitude ratio a 1 a 2 is weak.

[0057] Due to convolution, two closely spaced echo pulses interacting with a broad impulse response become entirely encompassed within that impulse response simultaneously, making it impossible to distinguish between them. Therefore, a conventional lidar L performing standard processing at the output of the amplification circuit is unable to differentiate between two nearby obstacles within its field of view.

[0058] There figure 5 This diagram synthesizes the previous analysis and illustrates the temporal evolution of different signals of interest. The initial pulse li is emitted at t=0, the pulse Ir1 (post) is received at R1, and the pulse Ir2 (vehicle) at R2. The signal s01(t) corresponds to the (noise-free) response of the amplification circuit to the presence of the post alone, the signal s02(t) to the (noise-free) response of the amplification circuit to the presence of the vehicle alone, and the signal s0(t) to the response of the amplification circuit to the presence of both obstacles, taking into account measurement noise. From s0(t) onward, it is impossible to discriminate between the two obstacles.

[0059] The lidar processing unit 10 according to the invention implements a method 100 for processing the received signal, enabling the resolution of the aforementioned problem of obstacle non-discrimination, illustrated figure 6 .

[0060] According to another aspect, the invention relates to a method 100 of lidar signal processing, which is applied to a lidar mounted on a static or moving object, and allows the detection of obstacles in front of the lidar.

[0061] It includes a first step A of digitizing the amplified electrical signal s0(t). We note s 0 e t This digitized signal, and Fe the sampling frequency. A digital system has the particularity, compared to analog systems, of only knowing information at the moments of sampling. T e = 1 F e . . The designation of digital information is indicated here by the superscript e for the signals resulting from the different processing.

[0062] In step B, at least one time-domain correction function, called the correction filter Ce<(t), is applied to the digitized amplified electrical signal. s 0 e t , in order to generate a processed signal s p e t Preferably, the application of the correction filter consists of a convolution of the digitized amplified electrical signal. s 0 e t with the time-correction function C(t), also digitized: s p e t = s 0 e ∗ C e t

[0063] The signal s0(t) at the output of the receiving device is conventionally determined by a convolution of the input light pulse Ir(t) with the impulse response hr(t) of DR (formula (5)). The measurement and processing chain is illustrated. figure 7 .

[0064] The filter C e< (t) acts in the time domain, directly on s 0 e t There is no transformation to move to Fourier space. The applied filter Ce(t) aims to improve the temporal resolution of the detection to discriminate obstacles. Ce(t) is determined from the sampled impulse response hre(t) and a predetermined (desired) time-domain analysis function hce(t). It is as if the response hr(t) had been replaced by a corrected impulse response hc(t), thanks to the corrector C(t), as illustrated. figure 8 and formula (7) below. The introduction of a discontinuity in the analysis function (see below) aims to obtain a corrected impulse response with a slope break allowing the discrimination of obstacles. hc t = hr * C t

[0065] For the implementation of method 100 according to the invention it is therefore necessary to have knowledge of the impulse response hr(t) (measurement and / or simulation), which is digitized and stored.

[0066] The transformation of the initial impulse response hr to the desired impulse response hc is carried out using a time-domain method.

[0067] The various calculations are, of course, performed on the digitized values ​​of these functions, as illustrated. figure 9 We call: h0, h1, h2 ... the different digitized values ​​of the impulse response hr(t) at sampling times t0, t1, t2... c0, c1, c2 ... the different digitized values ​​of the correction function C(t) at sampling times t0, t1, t2 ..., a0, a1, a2,... the different digitized values ​​of the analysis function hc(t) at sampling times t0, t1, t2 (in this example the analysis function has two discontinuities of values ​​a6 and a7 at two successive sampling times t6 and t7, the other values ​​being zero).

[0068] The correction filter C is determined by deconvolution of the impulse response hr by the analysis function hc: C t = hr * − 1 hc t

[0069] Applied to the digitized values, the coefficients ck of the compensator are determined using the following formulas: c 0 = a 0 h 0 c k ≥ 1 = 1 h 0 a k − ∑ m = 1 n h m ⋅ c k − m

[0070] Note that calculating the correction coefficients requires dividing by h0, which can pose numerical problems when h0 has a very small or even zero value. One solution is to add a constant to the entire impulse response and calculate the coefficients. The value of the constant is increased through successive iterations until convergent values ​​for the coefficients are reached.

[0071] An example of these coefficients ck is illustrated figure 10 .

[0072] The predetermined analysis function identified by the inventors has at least one non-zero value A0, called a discontinuity, at a given instant called the discontinuity instant td0, with a return to substantially zero values ​​around said discontinuity. The return is preferably fast. The analysis function is digitized at sampling points (see figure 9 ) and by rapid return to near-zero values ​​we mean a decrease that occurs over a small number of sampling points. The maximum number of points over which the return to zero must take place depends on Fe, the distance between objects, the amplitude of the echoes... In short, the separation performance is better when the decrease occurs over a small number of sampling points, but a gain always exists, however small, as long as the analysis function has one or more slopes greater than that(s) of the impulse response.

[0073] Thus, the important point is that this decrease acts as a discontinuity (break in slope) with respect to the slow variation of hr(t). Typically, the descent occurs over at most a few points to about ten sampling points.

[0074] Finally, in step C, a distance di from the element Ei (obstacle indexed i) is determined from the processed signal. s p e t This determination is carried out by a process 70 which consists of locating the local maxima(s) of s p e t and the instant(s) tmi corresponding to this or these local maxima. This temporal location of the local maximum tmi corresponds to the delay or flight time Ri for the associated obstacle Ei, tmi=Ri, and we deduce the distance di of the obstacle from this instant tmi, with formula (1).

[0075] Local maxima are sought within a time range of interest, depending on the application and / or the specific data being analyzed. Particular attention is paid to the obstacle closest to the object carrying the lidar and its distance from the lidar. The distance of interest within which the scene in front of the lidar is scanned depends on the application.

[0076] For example, the figure 11 shows the effect of a Gaussian-shaped analysis function hc(u) with a single discontinuity (a discontinuity is defined as a maximum with a return to values ​​close to zero) on a signal Ir comprising two pulses Ir1 and Ir2. The different curves A to D illustrate different times during the convolution. t = t 1: none of the impulses yet interact with the corrected impulse response hc. t = t 2: The impulse Ir1 interacts with the corrected impulse response h c but not the Ir2 impulse t = t 3: The impulse Ir1 no longer interacts with h c resulting in a net decrease in common areas and therefore in the output signal t = t 5: In turn, the impulse Ir2 no longer interacts with h c , causing a further sharp decrease in the output signal

[0077] Curve F illustrates the result of the hc * Ir convolution treatment: hc * Ir = Ir * hc = Ir * hr * C = s 0 * C = s p

[0078] The introduction of at least one discontinuity makes it possible to dissociate at some point the interactions of the two impulses with the impulse response, and makes it possible to clearly discriminate the times of arrival of the echoes.

[0079] Thus the lidar according to the invention makes it possible to detect a small obstacle in front of a larger obstacle, for example a smooth wall at an angle.

[0080] The lidar according to the invention offers advantages unavailable to ultrasonic sensors and can therefore replace them in collision avoidance systems for robotics. It exhibits improved characteristics in terms of: Sensitivity to weather conditions: insensitive to rain or high humidity; detection field: width on the order of 5°, 10°, 20°, or 30°; detectivity: detection of thin or low-reflectivity objects; obstacle separation: discrimination of close obstacles, detection of a thin object placed in front of a wide object, detection of multiple nearby echoes, despite a low bandwidth of the amplification circuit; acquisition: flash lidar, "single-shot" system possible (one laser shot = one result); refresh rate: high, greater than 10 kHz

[0081] Designed for portable embedded applications, it also offers interesting features in terms of integration: Reduced consumption (mono): power supply possible on batteries or accumulators; reduced dimensions: a few cm³ are possible; reduced mass: a few grams are possible, reduced cost.

[0082] It is adaptable: adding optics increases its range, while retaining echo discrimination capabilities.

[0083] These characteristics allow the use of lidar in the fields of robotics, drones, autonomous movement, collision avoidance...

[0084] The detection of multiple nearby echoes allows, among other things: to offer better knowledge of the surrounding terrain and allow better anticipation of the trajectory, the detection and consideration of weak echoes due to small obstacles placed closer than large obstacles, to improve the movement of a drone or robot in a dense overall context, for example in urban, forested areas.

[0085] The signal processing according to the invention, allowing the discrimination of multiple echoes with little computing power, may be of interest for other equipment measuring flight times, in the fields of reflectometry, sonar, ... and for the surveillance of premises.

[0086] The single-shot aspect also allows for greater discretion of the lidar.

[0087] At a sampling frequency F e corresponds to a sampling period T e , The quantum of time after which the digital system recalculates its information. The sampling period corresponds to the temporal resolution of the measurement, or to the spatial discrimination resolution in the case of lidar.

[0088] At the current temporal resolution T e corresponds to a spatial resolution of T e 6.67 ns In practice, the sampling frequency is determined by the reference of the analog-to-digital converter (ADC) used and defines the spatial discrimination resolution between obstacles. For example: 75cm @ F e = 200 MHz 37.5cm @ F e = 400 MHz 5.5cm @ F e = 3000 MHz

[0089] By generating one or more discontinuities in the impulse response of the processing chain, the discrimination power is no longer linked to the bandwidth of the receiver but to the sampling frequency of the digital system: the signal processing according to the invention makes it possible to use a low bandwidth receiver for its sensitivity and S / N (signal-to-noise) ratio qualities while benefiting from a high discrimination capacity.

[0090] For better sensitivity, at least one instant of discontinuity td0 is preferentially chosen in the vicinity of the instant of the maximum of the impulse response. h r ( t ), denoted t m-imp. We mean by neighborhood of t m-imp an instant located in a time interval around t m-imp such that the amplitude of the impulse response hr associated with this instant is greater than or equal to a non-zero fraction of the maximum amplitude of hr, preferably greater than or equal to the maximum amplitude divided by 4. Preferably at least one instant of discontinuity td0 coincides with t m-imp.

[0091] There figure 12 illustrates a limitation of using an analysis function that only has one discontinuity with a given width. The curves on the left are equivalent to the curve of the figure 5 The curves on the right illustrate the convolution with the unprocessed impulse response hr. Curves A and B correspond respectively to two impulses Ir1 and Ir2 separated by two different times, and more precisely, to obstacles that are further apart for A and closer together for B. When the echo impulses are sufficiently separated (relative to the width of the Gaussian), good separation is obtained (case A). For case B, the separation is of average quality. An analysis function of a certain width (Gaussian type in the example) can suffice to achieve obstacle separation, but, due to its non-zero width, it is less efficient than a Dirac-type function (infinitely narrow discontinuity).

[0092] The inventors identified three types of analysis functions that are particularly relevant when used in combination.

[0093] The first type, illustrated figure 13 and already discussed, is an analysis function exhibiting a single discontinuity (A0, td0). The function A on the left corresponds to a non-zero value A0 and a return to zero values ​​at a sampling point (cross) on each side.

[0094] The best results were obtained with a function such as that represented in B on the right, which has zero values ​​outside the discontinuity A0, also called the simple Dirac discontinuity. The discontinuity is then as steep as possible. Preferably td0 = t m-imp .

[0095] The second type, illustrated figure 14 is an analysis function with two discontinuities, a first discontinuity A1 at a first instant of discontinuity td1 and a second discontinuity A2 at an instant of discontinuity td2, td1 and td2 being close in time, that is to say separated by a few sampling points, because here it is a matter of separating close echo pulses.

[0096] The function A on the left corresponds to non-zero values ​​A1 and A2, returning to zero at a sampling point on each side, for both discontinuities (as an example). Again, as an example, there are 3 sampling points between times td1 and td2. Preferably, for better function efficiency, we choose two discontinuities of opposite signs: one with a positive value and one with a negative value. In this example, A1 > 0 and A2 < 0. For example, we place td1 at tm-imp, but this could be td2, or tm-imp could be located between td1 and td2. The important thing is that td1 and td2 are located in the vicinity of tm-imp.

[0097] The best results were obtained with a function such as the one represented in B on the right, which has zero values ​​outside discontinuities A1 and A2, separated by the minimum possible distance of no sampling points, also known as a double Dirac delta function. Here, the return to zero values ​​occurs in less than one sampling point, and the slope between the two discontinuities is as steep as possible. For example, td1 is placed at t m-imp, but it could also be td2.

[0098] The third type is an analysis function exhibiting three discontinuities close together in time.

[0099] According to one variant, the inventors have shown that these types of functions allow for improved processing when combined. This is made possible by the simplicity of the computational process. Thus, according to this variant, a plurality of correction filters Cj(t) (j being the filter index), precalculated from a plurality of predefined analysis functions hcj(t), are applied in order to generate a plurality of complementary associated processed signals sfj(t).

[0100] The distance of the elements of scene Ei is determined from the plurality of signals processed, by detecting local maxima and comparing their respective positions in the different signals processed.

[0101] An example is the detection of the obstacle closest to the lidar. In one embodiment, the first type of analysis function is used to identify the time range of interest in which this closest obstacle is detected. Then, the application of correction filters determined from the second type of analysis function, with different values ​​for A1 and / or A2, allows the final localization of this closest obstacle within the time range of interest. This enables the local detection of echoes, for example, very weak echoes masked by larger ones. Indeed, the inventors have shown that an analysis function with a single discontinuity allows for a complete scan of the space of interest (but may lack precision), while an analysis function with two discontinuities performs a "magnifying glass" function on the vicinity of the closest obstacle (see example below).

[0102] For example, when A2=-A1, the analysis function presented is equivalent to a derivative, commonly used for local studies of functions or signals. The result is that this analysis function will provide the derivative of the sum of the echoes, offering additional information about their location.

[0103] According to another example, the function exhibiting three discontinuities, for example equal to +1, -2, and +1 respectively, is equivalent to a second derivative.

[0104] Of course, the three types of Dirac mono, bi, and tri functions described below are only examples of possible analysis functions. Other pulse-specific analysis functions can also be considered to refine echo detection.

[0105] According to one variant, the plurality of corrective filters is applied via an iterative process, as illustrated figure 15 until a final processed signal allows the determination of a distance corresponding to the obstacle being sought, typically the nearest obstacle. In this case, the filters are determined progressively within the loop by recalculating coefficients c0, and / or c1 / c2, and / or c1, c2, and c3, based on the result obtained, i.e., the processed signal spj(t), and more specifically, the position of the local maxima(s). The iterative process consists of modifying discontinuities, i.e., non-zero values ​​of the aforementioned analysis functions. An optional branch 15 allows the measurement of the impulse response of the receiving device. According to one embodiment, this measurement is performed regularly during the implementation of the method according to the invention.

[0106] The iteration loop stops when the temporal location of the maximum of the obstacle of interest, here the closest one, no longer varies, which means that this obstacle has been successfully separated from the others.

[0107] According to one embodiment of the iterative process, a corrective filter corresponding to an analysis function with a single discontinuity A0 is first applied, then analysis functions with a first and second discontinuity (A1, A2) at two discontinuity instants (td1, td2) close in time, and / or analysis functions with three discontinuities at three discontinuity instants close in time, the discontinuities being modified iteratively.

[0108] Simulations were performed based on an example of lidar implementation according to the invention. Experimental measurements were also carried out, with experimental results close to the simulations, demonstrating the validity of the calculation. The goal here is to determine the distance to the nearest obstacle in a scenario with three obstacles in a first configuration. Obstacle 1: d=0.7 m - amplitude: 1 Obstacle 2: d=1.7 m - amplitude: 1 Obstacle 3: d=6 m - amplitude: 5

[0109] The simulation parameters are as follows: Sampling frequency Fe = 400 MHz (one point every 2.5 ns), 10-bit sampling Amplification circuit: R = 300 kΩ Bandwidth: 2.7 MHz The time extension of hr(t) is 200 ns or 80 sampling points.

[0110] There figure 16 illustrates the raw and processed lidar signals for three situations: (1): Obstacle 1 only, (2): obstacles 1 and 2, (3): obstacles 1, 2 and 3 in the scene.

[0111] Lines D1, D2 and D3 represent the exact position of the three obstacles that are behind schedule.

[0112] The curves on the left illustrate the raw signal s0(t) at the receiver output, as shown. figure 5 The curves on the right illustrate the processed signal spe<(t), with an abscissa corresponding to the time quantums (sampling points) i.e. 2.5 ns (which corresponds to a distance quantum of 2.5 / 6.67 = 37.5 cm).

[0113] The position of the first obstacle is at abscissa 10, which is a delay of 25ns. This does not directly correspond to the position of the first obstacle, because there is a time offset, perfectly controlled, which depends directly on the position of the Dirac delta function of the analysis function.

[0114] The simulation is performed using a simple Dirac first-type analysis function hc, as illustrated. figure 17-A : a single discontinuity, of amplitude A0=1 for a sampling point at a time td0 = 20ns, and zero values ​​for the other times.

[0115] We note that in this obstacle configuration the analysis function of the first type allows the identification of the 3 obstacles and their separation.

[0116] There figure 18 illustrates the signal processed with the same analysis function as before for the following second obstacle configuration: Obstacle 1: d=0.7 m - amplitude: 1 Obstacle 2: d=1.7 m - amplitude: 10 Obstacle 3: d=6 m - amplitude: 5

[0117] The echo from obstacle 2 is much stronger here. We can see on the curve that the peak associated with the first obstacle is still visible, but weaker. For obstacles that are too close together and / or with a second, strong obstacle, we observe the limit of the simple Dirac delta function.

[0118] There figure 19 illustrates the signal processed with the same analysis function as before for a third obstacle configuration: Obstacle 1: d=0.7 m - amplitude: 1 Obstacle 2: d=1.2 m - amplitude: 10 Obstacle 3: d=6 m - amplitude: 5

[0119] Obstacle 2 has been brought closer to obstacle 1. We note that the first two obstacles are no longer discriminated by the simple Dirac analysis function, and that the processed signal corresponds to a fictitious obstacle located between the two real obstacles.

[0120] In order to solve this problem and accurately identify the position of the nearest obstacle, a double Dirac analysis function is then applied, preferably according to an iterative process.

[0121] There figure 20 Figure A illustrates the signals processed with the Dirac dual analysis function for the third obstacle configuration. Figure B shows the different processed signals obtained for various values ​​of a6, which is increased by 1 at each iteration. It can be seen that between the first two iterations (X=1 and X=2) and the third iteration (X=3), the position of the first obstacle is not stabilized and varies. It is therefore necessary to continue the iteration.

[0122] Then for the following iterations this position becomes stable; Once the position of the first obstacle is stabilized, the process can be stopped.

[0123] Note that when applying the double Dirac delta function, only the position of the first obstacle should be considered. What happens after (in time) should not be taken into account.

[0124] According to one variant, the AC amplification circuit of the lidar 10 according to the invention has an additional component.

[0125] The most suitable and commonly used receiver circuit associated with a PD photodiode is classically called a transimpedance amplifier (" Trans Impedance Amplifier "or TIA. Its components are known to professionals. In its most basic version, it consists of a resistor" R f and an operational amplifier (Amp) as illustrated figure 21 .

[0126] Considering the ideal components: The PD photodiode transforms the light flux into photo-generation current I ph . The TIA transforms the current I ph in tension according to the relationship S = - R f I ph . The resistance value R f sets the amplifier gain. The sensitivity of the TIA is related to this value.

[0127] A PIN type photodiode is preferred, being more reliable and simpler to implement than an avalanche photodiode (APD).

[0128] The current from the photodiode can be described by the following relationship: i ph t = S . ∅ e t + I 0 + i n ph t In which: S is the sensitivity of the photodiode, on the order of 0.6 A / W, Ø e ( t ) is the received light power, I 0 is the sum of the inverse static currents of the photodiode (saturation current, dark current), i nph ( t ) represents the sum of the intrinsic noises of the photodiode (mainly shot noise).

[0129] The light power captured Ø e ( t ) consists of a dynamic part φ e ( t) such as the reflected impulse and a static part caused by a luminous background Ø e 0, caused by the sun for example (see also formula (3) and A0). The relationship can then be written as: i ph t = S . φ e t + ∅ e 0 + I 0 + i n ph t

[0130] A PIN photodiode generates approximately 600 mA / W. The transimpedance amplifier has a gain that is generally a compromise between the desired bandwidth and the sensitivity (detection capability) of the sensor. It may also be limited by the intensity of the photodiode's reverse static currents.

[0131] In terms of sensor sensitivity and measurement signal-to-noise ratio (S / N), a resistance should be used. R f high leading to high gain (and therefore low bandwidth).

[0132] Ambient light is amplified in the same way as the optical signals of echoes. The orders of magnitude are very different between the currents generated by the echo pulses (typically a few µA for an echo of a few µW ) and that induced by ambient light (typically a few tens of mA for a 5×5mm 2< silicon photodiode).

[0133] To limit the photogeneration current induced by the sun, which can lead to saturation of the photodiode's optical current array (PDA), a conventional optical filter is placed directly on the surface of the photodiode. This can be the color filter integrated into the photodiode by manufacturers (broad spectrum on the order of 300 nm), or an interference filter (narrow spectrum on the order of 10 nm – which has a major drawback of limited directivity).

[0134] However, despite the presence of a filter, an output voltage S higher than the supply voltage of the TIA (typically 5V) is quickly reached with a high resistance Rf and average ambient illumination (sunlight), which leads to saturation of the sensor.

[0135] Therefore, the sensor's performance in direct sunlight is incompatible with the use of a high-sensitivity TIA. A compromise on gain is necessary: The advantage of a high gain (TIA resistance) R f of high value) to obtain a high S / N ratio necessary for detecting obstacles that return few echoes; The preference for low gain (TIA resistance) R f of low value) in order to prioritize high bandwidth for obstacle discrimination; The preference for low gain (TIA resistance) R f of low value) for the sensor's resistance to full sunlight.

[0136] These opposing points demonstrate the difficulty in achieving a wide-field proximity lidar capable of discriminating between different obstacles included in the detection field and those close to it, characterizing their distances, and maximizing the chances of detecting the nearest obstacle.

[0137] To solve the problem of sensor stability in direct sunlight, the lidar receiver DR of the 10 according to the invention includes a transformer T and a capacitor C between the photodetector PD and the amplifier TIA, as illustrated. figure 22 The primary of the transformer is connected to the anode of the photodetector, the secondary is connected to the capacitor C, this capacitor being connected to an input of the transimpedance amplifier.

[0138] The transformer can be mounted in a step-down configuration (Vout<Vin, Iout> (in) or in elevator (Vout>Vin, Iout <Iin). Préférentiellement le transformateur est monté en abaisseur ce qui permet d'amplifier le courant de photodiode du rapport de transformation sans ajout de bruit complémentaire.

[0139] The operating principle of the transformer lies in the conversion of a current in its primary (here the photodiode current). i ph ( t ) = S. ( φ e ( t ) + Ø e 0) + I 0) in a magnetic field B ( t ), itself converted into an electric field E ( t ), therefore in a tension s t ( t ) (Vout) to the secondary of the transformer.

[0140] It is known that the electric field is a function of the derivative of the magnetic field induced by i ph ( t ) .Consequently, the voltage and current from the transformer depend only on variations in i ph ( t ) . The voltage resulting from the TIA will be of the form: V s t = k . d dt φ e t

[0141] Without characterizing the coefficient k It is established here that all the static components of the photodiode current i ph ( t ) are eliminated at the secondary side of the transformer: The intrinsic reverse parasitic currents of the photodiode (saturation current, dark current) The photogeneration currents due to any light sources, weak or intense (sun).

[0142] The capacitor C placed between the TIA and the transformer serves to minimize the gain the system would have without this capacitor. Indeed, this assembly would behave as a non-inverting amplifier with infinite gain relative to the op-amp's offset voltage, resulting in its saturation.

[0143] The structure of the DR receiving device according to the invention therefore solves two problems: The gain resistance limit is lifted because various DC parasitic currents are filtered out. A very high-gain sensor is created, and the detection capability is improved: the sensor is virtually insensitive to ambient light. It is then possible to detect obstacles even when facing the sun.

[0144] This structure is perfectly suited for use with a wide lidar F.O. V outdoors.

Claims

1. A method (100) for processing a signal from a lidar, said lidar performing a time-of-flight measurement and comprising an emitting device configured to emit light pulses in the direction of a scene at an angle greater than or equal to 5° and a receiving device, said receiving device exhibiting an impulse response (hr(t)) and comprising a photodetector configured to receive pulses reflected or backscattered by at least one element (Ei) of the scene and to convert said pulses into an electrical signal, and an amplification circuit (CA) configured to generate an amplified electrical signal (s0(t)), the method comprising the steps of: ∘ A: digitizing the amplified electrical signal ( s 0 e t ) ∘ B: applying at least one time correction function, referred to as the correcting filter (Ce(t)), to the digitizing amplified electrical signal in order to generate a processed signal (sf(t)); applying the correcting filter consists in convolving the digitized amplified electrical signal with said correction time function, the correcting filter (Ce(t)) being determined based on the impulse response and a predetermined time analysis function, the analysis function having at least one non-zero value (a0, a1, a2), referred to as the discontinuity, at a given time referred to as the discontinuity time (td0, td1, td2), with a return to substantially zero values around the discontinuity; said correcting filter is determined by deconvolution of said impulse response by said predetermined analysis function, ∘ C: determining a distance (di) of said at least one element (Ei) based on the processed signal.

2. The processing method according to any one of the preceding claims, wherein a presence of said at least one element in said scene corresponds to a local maximum of said processed signal, and said associated distance is determined from a temporal location of said local maximum.

3. The processing method according to one of the preceding claims, wherein said impulse response has a maximum at a time tm-imp, and wherein said at least one discontinuity time of the analysis function temporally coincides with said time tm-imp.

4. The method according to one of the preceding claims, wherein the analysis function has zero values outside said at least one discontinuity.

5. The method according to one of the preceding claims wherein the analysis function has either a single discontinuity (A0), or two discontinuities, or three discontinuities, located respectively at discontinuity times close together in time, i.e. separated by a few sampling points.

6. The processing method according to one of the preceding claims, wherein a plurality of correcting filters (Cj(t)) determined from a plurality of analysis functions (hcj(t)) are applied, so as to generate a plurality of associated processed signals (sfj(t)), said distance of said at least one element in the scene being determined from said plurality of processed signals.

7. The processing method according to the preceding claim, wherein said plurality of correcting filters is applied via an iterative process, until a final processed signal allows the determination of a distance corresponding to the nearest obstacle; the iterative process consisting in modifying discontinuities, that is non-zero values of said analysis functions.

8. The processing method according to the preceding claim, wherein a correcting filter corresponding to an analysis function with a single discontinuity (A0) is first applied, followed by analysis functions with two discontinuities or three discontinuities, said discontinuities being iteratively modified.

9. A time-of-flight lidar system (10) comprising: - an emitting device (DE) configured to emit light pulses towards a scene at an angle greater than or equal to 5° - a receiving device (DR) having an impulse response (hr(t)) and comprising: ∘ a photodetector (PD) configured to receive pulses (Ir) reflected or backscattered by at least one element (Ei) in the scene, and to convert said pulses into an electrical signal, ∘ an amplification circuit (CA) configured to amplify said electrical signal, - a processing unit (UT) of said amplified electrical signal configured to: ∘ digitize the amplified electrical signal ( s 0 e t ) ∘ apply at least one time correction function, referred to as the correcting filter (Ce(t)), to the digitizing amplified electrical signal in order to generate a processed signal (sf(t)); applying the correcting filter consisting in convolving the digitized amplified electrical signal with said temporal correction function, the correcting filter (Ce(t)) being determined based on the impulse response (hr(t)) and a predetermined time analysis function, the analysis function having at least one non-zero value (a0, a1, a2), referred to as the discontinuity, at a given time referred to as the discontinuity time (td0, td1, td2), with a return to substantially zero values around the discontinuity; said correcting filter is determined by deconvolution of said impulse response by said predetermined analysis function, ∘ determine a distance (di) of the at least one element (Ei) based on the processed signal.

10. The lidar system according to the preceding claim, wherein the amplification circuit (CA) comprises a transimpedance amplifier (TIA), a transformer comprising a primary and a secondary, and a capacitor (C), the primary of the transformer being connected to an anode of the photodetector, the secondary being connected to a capacitor, said capacitor being connected to an input of said transimpedance amplifier.

Images